• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Early AMD K10 Scores

AMDPower

AMDPower

AMDPower

Soldato
Joined
24 Feb 2004
Posts
3,028
saw these over at extremesystems forum

zmhjxx0wenmlwemyzmzk.jpg


it is not looking good for AMD:o

http://www.xtremesystems.org/forums/showthread.php?t=157136
 
Pinch of salt with those, my old X2 4400+ at 2.4 did it in 30s, so I'd at least expect it to match that.

I think until it gets close to release, we're going to see all sorts of results from one extreme to the other, its upto those who look at them to decide which they beleive
 
Would be good if there was a multithreaded version of SuperPI :o

Also SP loves memory bandwidth, so that result depends a lot on the test system. And because it only uses one core, it's only getting 512KiB of L2 cache available to it.

Jon
 
Phenom's largest gains may well come from applications which make full use of all 4 cores, as the single L3 cache will help keep the cores flowing.

On single threaded applications, they will be relying on other improvements, like the rumoured twin FPU's per core. But that depends on what the program needs. Im guessing there will be benchy's which show the Phenom to be very fast, and others showing Core 2 is still very good. In the real world.. who knows Perhaps Phenom will be fastest, or Perhaps it will be Core2.. Or it could be so close, that its impossible to call it.

We'll only know for sure when the production chips are out. But I hope for AMD that they already have the next chip's ready to go to test silicon.. Intel are far down the road towards Nehelem, and are probably already working on the next chip after that as well.

It seems likely that AMD and Intel will now alternate as to who has the fastest chip at any given 'stock' speed, but Intel wont just sit back on a single core design, trying to make a single design 'work'. P4 should have been scrapped years ago, but they continued to try and fix it, giving AMD a long time ruling the top performance.
 
Take these with a bucket load of salt tbh, plenty of people want the increased traffic, this is the good way to do it. Same as the recent K10 pwns 3D'06 news, until I see the evidence then there is no way I could believe it.
The way things are going you'd think people just build pc's purely so they can run SuperPi and touch themselves when they get a low time... Actually, I believe that probably does happen...
 
is phenom going to be TRUE quad core, or quad core like old pentium D's were dual core, and new q series processors are quad...?
 
something just isn't sitting right with me on the benchmarks. the cine 3d benchmark just seems to me that something is different in the setups.
Its just that they hide certain aspects of the intel and the amd images. IIRC cine3d can make use of firegl/quadro cards so this in itself would speed up rendering performance especially if its missing from the amd system.
And iirc superpi has always been an intel domain for the results anyway.

I'll wait till the final release to pass overall judgement.
 
tbh, I could post some edits of my duron 1.1ghz benchmarks and stick a Phenom CPU-Z screen in and they probably be lapped up by the a lot of people...
To the extremes, but the way some of the paralysed mind fanatics are going about it on xtreme you'd think the world had ended and these results were gospel truth.
 
Hmm, Do you think they will really call it SSE5, I wonder if Intel will also call their next revision of SSE SSE5, Why couldnt AMD have just called it something else, like they did with MMX/3D-Now.

Intel were behind SSE/2/3/4. Guess Nehelem will have to skip to SSE6 if there are any improvements planned in the SSE on that platform. Penryn of course adds SSE4.
 
I guess now we know why AMD has kept the benchies close to its chest for over a year then.

:(

Next stop, Nehalem!
 
NathanE said:
I guess now we know why AMD has kept the benchies close to its chest for over a year then.

:(

Next stop, Nehalem!
Click to expand...

Kinda jumping the gun a bit on that one I think like... wait till the chips are actually out and we get some "reputable" reviews. ;)
 
You must log in or register to reply here.
Back
Top Bottom